In clinical oncology and drug development there is genuine need for sensitive and reproducible quantitative imaging methods for early prediction of therapy response and accurate assessment of post-therapy residual cancer. Such methods have the potential to spare non-responding patients from ineffective and/or toxic treatments, improve clinical management, and accelerate efficacy evaluation of novel therapies. Dynamic- Contrast-Enhanced (DOE) MRI can provide an excellent measure of therapy-induced tumor vascular changes. The Standard-Model (SM) for pharmacokinetic DCE-MRl analysis incorrectly assumes effectively infinitely fast equilibrium inter-compartmental water exchange kinetics and, as a result, often underestimates the pharmacokinetic parameters Ktrans and Ve. The Shutter-Speed Model (SSM) accounts for finite water exchange kinetics effects and corrects the imaging biomarker underestimations. It has proven more sensitive to vascular changes than the SM. In addition, SSM DCE-MRl allows quantification of novel imaging biomarkers, such as ?Ktrans(= Ktrans (SSM) - Ktrans (SM)] - a measure of precisely the water exchange (shutter-speed) effect on Ktrans estimation. ?Ktransis not only a more sensitive biomarker for therapeutic response, but also less prone to other systematic errors often observed in DCE-MRl quantification. In Specific Aim 1, the Shutter-Speed Model will be applied to phase l/ll clinical trials in two disease areas (breast cancer and soft tissue sarcoma), SSM DCE-MRl will be compared/combined with SM DCE-MRl, diffusion-weighted MRI and tumor size measurement for assessment of therapy response. In Aim 2, the effects of data acquisition and processing schemes on DCE-MRl biomarker values will be investigated within the context of therapeutic monitoring. In Aims 1 and 2, pathology results will be used as endpoints for correlation with imaging results and statistical analyses. In Aim 3, caBIG-compliant software tools will be developed that utilize a single and/or a set of imaging biomarkers to aid clinical research.